Facebook is, allegedly, an amazing company with some of the best developers ... that don't have a basic check if a new profile has the same name and profile picture of an existing one that is clearly fraudulent.
They probably only investigate and close imposters if they receive complaints from other users.
Perhaps machine learning would be able to identify fake accounts with a higher degree of certainty than the 'matching name/picture' check, and only fakes with a high certainty rate are deleted.
From a different angle though, if no one has reported a fake account - does Facebook care about the existence of a fake account? Some friends a while ago created a fake account of a meme amongst our group. Funny, and not harming anyone. If Facebook blocked this account it would seem a little domineering/controlling. I'm not sure that's the impression Facebook wants to leave on people who make these sorts of accounts for a laugh.
Until the systems can sort this stuff out, these things are simply going to be an issue.
It isn't even that the profiles are necessarily fraudulent. I know there are plenty of young (and older) occultists that keep two seperate accounts, in part to be able to keep that part of their life private and away from extended family and work. This happens in other circles as well.