Can I have that as a title?
The problem with a brain transplant is twofold; firstly the brain is reliant on millions of connections to the spine and each of these connections is measured at a cellular level. Because of this it is pretty much impossible to envision how these connections can be severed, the brain held in a stable condition and then reconnected in another body. Secondly the brain is the seat of the mind and therefore any change to it (i.e. replacing chunks) will damage the personality inside even if you could regenerate the connections perfectly.
This is the field of http://en.wikipedia.org/wiki/Regenerative_medicine" in question and control cell behaviour in desirable ways. The ideal scaffold would be one that could be placed in a patient, attract stem cells from the body, induce their differentiation whilst coordinating millions of cells at once to regenerate multiple tissue types and finally dissolve (or technically dissolve as it goes).
The biggest obstacles to these is understanding the various ways in which cells respond to environmental stimuli. Cells are highly sensitive entities; slight changes in pressure, pH, mechanical stimulus, chemical concentrations etc can have huge effects. So at the moment a lot of regenerative medicine research focuses on discovering what effects different conditions have for example: certain neurons extend their axons in parallel to deep groves and perpendicular to shallow ones. This lack of knowledge and lack of sophisticated technology mean that regenerative medicine products are limited to simple tissues rather than whole organs.
Other obstacles include the difficulty in creating a scaffold that can coordinate multiple cell types via mechanisms that don't interfere (or complement) with the other types of cells and angiogenesis. The latter is a big problem because as it stands
in vitro tissues are limited in size because they have no vasculature and so must rely on oxygen diffusion for respiration.
