Is this related to the recent increase in available close votes per user?
Without a doubt.
Is this intentional?
Sure, questions don't get into the review queue by winning a lottery. The votes to close are cast with an intent, which is to close the question.
Were all these questions in the system already, waiting for months for review?
No, Math.SE does not have a long backlog like Stack Overflow. Questions get reviewed within several hours.
are these simply because there are a great number of people using their 50 votes per day on old questions?
It does not take a great number of people. A small, but nonzero number of people is enough.
many people will spend their entire 20 reviews looking at months-old questions
Only if they choose to. A reviewer not wishing to engage with an old post can click Skip without reading the question. The age of question is shown in bold font in the sidebar.
Another option is to use the bookmarklet "Skip but open in another tab". When the question is voted on in another tab, the vote does not count toward the limit of 20 reviews. Combined with one-click closing bookmarklets, this is a fairly efficient workflow for up to 50 closed questions per day.