In addition to addressing speciﬁc patterns of behavioral and neural data, neural network models are valuable for their ability to establish general principles of functional neural organization. In particular, computational models can explain how diﬀerences in the structuring and parameters of neural networks can lead to qualitatively diﬀerent, often mutually incompatible, capabilities. To the extent that diﬀerent computational capacities require fundamentally diﬀerent kinds of neural networks, the brain could have either a compromise or a trade-oﬀ between the diﬀerent network properties, or it could specialize diﬀerent brain areas for diﬀerent functions to avoid such a trade-oﬀ. Critically, this kind of computational approach to functional neural organization enables one to understand both what is diﬀerent about the way diﬀerent neural systems learn, and why, from a functional perspective, they should have these diﬀerences in the ﬁrst place. Thus, the computational approach can go beyond mere description towards understanding the deeper principles underlying the organization of the cognitive system.