"Neural computation" covers a wide range of activities. At one extreme are people who
spend all their time working on theory that draws a few inspirations from experimental
neuroscience. At the other extreme are experimentalists who use computational modeling
as just another research tool, like voltage clamping or genetic manipulation of channel
properties.
Instead of loading you up with an impossible list of difficult books, I suggest that you
start with "From Computer to Brain" by William W. Lytton. This well-written book will
introduce you to many key concepts. It is easy and fun to read. Also, the author uses
NEURON for the examples and exercises, and has posted the source code on the
WWW--see
http://www.neuron.yale.edu/fctb/