Thing about ASC, despite its extreme heritability, it's not only a spectrum within itself but part of the larger human spectrum. Meaning, the genetics underlying it are potentially common throughout the entire race's genome. Plus, concerning the specific neuroanatomical makeup of ASC, there are potentially different factors (genomic, nongenomic, and environmental) which can lead to similar ends.
So say you have a single neuroanatomical characteristic that, at a certain extreme, underlies autism. This phenotype has a long path through development (multiple stages during prenatal development, then early postnatal, then later postnatal, etc.) in order to arrive to where it is, the aspect which defines autism. Given this single path trajectory, there is the potential that multiple factors at different times could all divert said path, increasing the likelihood of autism. But they're not all the same causes.
When it comes to the development of the cortex, for instance, you can ultimately define total cell number by:
1) prolonging the division of progenitor cells (symmetric division in which progenitors continue to give rise to more progenitors) which will eventually end up giving rise to all the neurons;
2) prolonging the asymmetric cell division of progenitor cells as they give rise to new neurons;
3) disrupting apoptotic mechanisms so that pruning does not occur appropriately.
Each of these three mechanisms will increase total cell count, but they each have the potential to arise from different genomic/nongenomic/environmental factors.