Giving us some indication of what country you're studying in can be helpful for recommendations generally of this nature. Based on your phrasing I would guess you're in the UK or some other commonwealth country? Additionally, posting the same thread twice times in different forums is kind of redundant...that said, I've skimmed the responses across the two threads and come up with the following:
First of all, as micromass says, mathematics is essential. To apply to MIT I believe (unless they've changed their requirements recently) you need to take one of the science SAT subject tests, and one of the two mathematics ones, in addition to the SAT or ACT. You don't need to take these until the year you are applying in or the year before (probably year 11 or 12 depending on where you live) so they're a while off.
From a UK/IB qualifications perspective (as that's what I'm familiar with), the basic SAT is GCSE level but has a lot of peculiarities in how it's formatted (the whole standardized schtick) so I recommend trying the practice test on their website, and the SAT book may be useful. You can take the ACT as an alternative however it requires a higher level of mathematics than the SAT (specifically I think it covers matrices and complex numbers).
Several other posters have given you recommendations such as Young and Freedman. This is a standard first year university physics text (it's the one my uni uses for the Physics degrees), however based on what you've said I think you will struggle with it for now. Additionally even that alone is overkill for the undergrad entry test; it covers material (last I checked) similar to GCSE and AS/IB SL Physics in the UK. In the US AP Physics would cover the necessary material and then some. Similarly for the mathematics tests the non-calculus topics from AS level maths/further maths (you might need matrices and complex numbers but I'm not sure) or IB SL maths will be sufficient. You just need to do a lot of practice problems to ensure you're comfortable with it; it's not inherently difficult mathematics and if you struggle with it, you will probably struggle with the first year physics and calculus classes at MIT (I have not taken them however, so I'm basing this on comparison to similar courses I've taken at different institutions, which were probably taught more poorly).
You should look at lower level physics and mathematics material to being self studying before moving onto anything like Young and Freedman. I would recommend starting with BBC Bitesize GCSE maths and physics as a refresher, you will likely be familiar with some of the material. Make sure to do the problems and review sections you get wrong. Having done this, look into A-level/IB maths and physics stuff on the web. I found
http://www.mathsrevision.net/advanced-level-level-maths-revision to be a pretty useful resource for high school level mathematics, it covers pretty much all your bases. Make sure you're completely confident with the GCSE maths content before moving onto that.
More generally however, you're far enough off that you don't NEED to buy a book and self study now for a test you'll be taking in probably 3 years time. By the time you get to that point, you'll have almost certainly covered at least most of the relevant material in your classes at school. When you're a few months away from taking the test THEN I might recommend cracking open young and freedman and having a go at some of the stuff in there. Trying to do it now will just distract you from doing well in your classes now (which will weaken your overall application, as you can't really just ace the SATs and get into MIT as I hear they're very into individuals experiences and suchlike) and may end up discouraging you in the long run if you get stuck. Particularly if you're struggling at chemistry and biology, you should be focusing on improving your work in those subjects, rather than trying to find something else. You have to take biology and chemistry courses in your first year at MIT I believe anyway, so you can't just avoid them forever :P
Also, look for fun things to do outside of your studies, that you're genuinely interested in. Places like MIT love seeing people who have gone out and done sciencey stuff just because they're interested in it. Don't try to do extracurriculars to tick boxes; make sure you have some genuine interest in it. If you're interested in astrophysics and astronomy, see what stuff you can do with that in your area. Are there any planetariums or observatories nearby? You could try to arrange meteor shower viewing parties at your school. A lot of universities, especially in the UK, are trying more and more to arrange outreach programs (especially in the sciences) for young students. I know at my uni (the University of Exeter), we have a student led project building a radio telescope in Cornwall, and we're trying to get local schools involved by doing talks with them and hopefully when the telescope is completed, have them come along and see it and explain it working in action. They don't necessarily have to be science based, although if you can this will help (as this is fundamentally what you're trying to get yourself into, it helps if you can demonstrate you're not going to burn out before the 4 years of your degree are over).
One guy from my high school, who went on to do Earth Sciences/Geology at Oxford, was very interested in environmental issues beforehand. He had done well in his GCSEs, did A-level Physics, Geography and whatever (maths I assume but idk he was a friend of a friend so I didn't talk to him much being an overworked IBer xD ) and in the process, over one of the summers went out to somewhere on the Jurassic coast and did some data gathering (no idea what the topic was, something to do with seawater measurements). He then wrote this up, entered into a competition and ended up winning the "UK Young Scientist of the Year" award (I mean it wasn't an accident he put a lot of work into it but he would've been happy just getting to go there and show his poster and stuff). This no doubt helped his application later...you're not going to be doing publishable original research, but you should try doing stuff in a scientifically rigorous way, recording the results and interpreting them.
(Also don't try to build a radio telescope from scratch it's cost us £7500 so far and we don't even have any money to pay the long term electricity bills left from all our grants xD )