Virtually all cortical systems feature “functional organization”: the arrangement of neurons with specific functional properties into characteristic spatial arrangements. Functional organization is ubiquitous across systems and species, highly reproducible, and central to our understanding of cortical development and function, but we lack a unified framework to explain its emergence and utility. In this talk, I’ll present the Topographic Deep Artificial Neural Network (TDANN), a computational model instantiating the hypothesis that cortical systems are optimized to balance representation learning with spatial smoothness constraints. I’ll demonstrate that the TDANN makes quantitatively accurate predictions of functional organization in the primate visual system, and in turn balances task performance with wiring length. I’ll also highlight our findings that 1) better topographic prediction translates to stronger prediction of neural responses by decreasing intrinsic dimensionality, 2) experimental manipulation of objective functions can yield surprisingly precise insights into the nature of cortical circuits, and 3) the resulting models can be applied practically to more directly interface with applied neuroscience.