I don’t play PF2E, but I checked out the crit rules a couple of months ago and rolling a nat 1 or 20 was still accounted for. From memory, doing that or succeeding / failing by 10+ brought you down or up a degree of success, and those degrees were:
Crit fail
Fail
Succeed
Crit success
So if the DC was 25, you rolled a nat 20, and your result was a 16-24, you would succeed. If your result had been a 25 or higher, you’d have critically succeeded.
If the DC was 10 and you rolled a nat 1, you didn’t necessarily critically fail. If your result was a 20 or higher, you’d still succeed (since you were over by 10); on a 10-19, you’d fail; and on a 9 or lower you’d critically fail.
If you rolled a nat 2-19, though, the impact of the over by 10 / under by 10 would be more noticeable. Vs a DC 15, you would crit fail if your result was 5 or lower and crit succeed if your result was 25 or higher.
Like I said, I haven’t played with the PF2E system, but my impression is that it would be a big improvement over 5e’s current system without suffering the same issues that most homebrew crit systems run into. It makes critical successes and failures more believable. It encourages DMs to prepare outcomes for the higher degrees of success / failure due to them having a higher chance of occurring. It makes it less likely for crit failures to happen to highly skilled characters.
And, as a fringe benefit, it also means that you can have almost impossible DCs that are possible only for the most skilled, and even then, only when they’re lucky (like a DC of 40 in a game where the highest bonus you can reasonably get is +11-+19, such that only a nat 20 by someone with such a bonus can succeed).