op 599: good for beginners, I respect Czerny very much, the only flaw is that there are too many studies in the same key. But apart from that, I think a beginner can progress quickly because each study is focused on a specific technique. I finished it quite quickly, but since I 'skipped' various things in the past (ehm...) I wanted to be sure that there aren't too many gaps in my technique.
op 139: I started practicing this today and I am very satisfied with it; this seems like the logical step after op 599: always easy studies of high quality but in many more keys, including minor keys.
op 821: the bomb. I like this collection very much, but the studies are quite more demanding. Real pieces of classical music concentrated into only 8 measures (!). Op 599 and 139, on the other hand, are longer structures, mostly binary. The beautiful thing about Czerny's studies is that no study is an 'indefinite project'; with a bit of application, you learn a lot and quite quickly. But in 8 measures? Even quicker than that!
In my opinion, this author is unfairly undervalued; I have heard a lot of nonsense about him, like saying he is similar to Hanon (yes, like saying a cow is similar to a house).
Aside from practicing scales in all keys for 30 minutes a day (which is already a lot for me, because I fall asleep afterwards), I have no intention of doing scale exercises without any musical sense, even though you certainly learn something from Hanon anyway. Why practice exercises without music when musical exercises exist?