Is it reasonable to suppose that perverts would be more likely to offend or just consume more porn?

Empirical studies have actually shown that consumers of child porn are unlikely to molest children. Those that are arrested for possession of child porn AND have a high likelihood of future molestation ALREADY had a history of molestation.

This shouldn't be surprising. Consumers of rape fetish porn are unlikely to actually rape someone, but actual rapists are likely to consume rape fetish porn. Furthermore, in countries where porn of all kinds was legalized, including child porn, enjoyed falling rates of child molestation.

The OP is making some of the same old arguments people have trotted out against porn for decades, just couched in pseudo neuroscience. The fact is, we don't understand these relationships to the degree the OP needs to justify his causal claims.

Maybe, maybe not. That's a fraught subject around pornography in general, with tendentious argumentation on every side.

On the other hand, children are extraordinarily vulnerable to many forms of abuse, this among them. In a system of laws one of whose explicit purposes is to afford those vulnerable to mistreatment protection under law, I don't think it is on its face unreasonable to argue that children merit extraordinary protection as well. While there is perhaps a fair question to be asked around whether the sort of law under discussion actually serves that end, I would at the very least suggest anyone raising the question in a serious way be very well prepared to answer objections and counterarguments of every imaginable sort.