OK, I think I see the problem...you have no idea what you are doing...
The "usual" brute-force way to do this is sketched in the drawing you provided. My guess is that the counters are daisy-chained the way I said before, output qD to the clock input of the next higher order counter. Then you need to have that "latch" between the counter outputs and the decoders to grab and hold the count on some periodic basis. Otherwise the displayed count will just keep spinning around before your eyes. Just after the latch data "hold" signal you need to reset all the counters so they start from zero again. That's the mystery line connecting them all and it needs to be driven by some regular periodic oscillator, so that you get, for instance, the actual count every second. You need to be careful that the latching happens before the counters are reset or all you'll get will be 0000's. So some subtleties abound here...
You might be able to find counters with built-in latches...I'm not sure as I haven't used much discrete logic in the last 20 years. By searching more, you may also be able to find better schematics for something that will actually work the way you want. I suspect that the frequency you are counting is not _way_ high, so you might have more success with CMOS logic (40xxx series) where there may be more options than good'ole 74xx TTL.
Another approach would be to use a PIC micro-controller to do all the counting and display driving, but that means learning to program along with everything else.
In any case I recommend that you get some parts and their data-sheets and a breadboard and start experimenting.
