A topic that I have been wondering about is what happens when two maps project fiber tracts onto a third - what is the plumbing and what exactly is calculated. How do the reverse connections work? I am thinking of how the feature extractors around V1 extract texture/stereo/motion/color and recombine these maps to refine edges in V1. I am pretty sure maps split and join all over the cortex. This goes to the H of HTM.