During recovery, how is the current sequence number used to determine the redo log files used to re-construct the database as well as the order in which to apply the changes recorded in them?
Perhaps a start on your journey to discovering this:
From the Oracle FAQs:
Every redo log file has both a log sequence number and low and high SCN.
Having said that, why do you feel you need to know this?
Suffice to say that Oracle (and RMAN in particular) does know this and you can use this to recover your database to any Point in Time. Does the actual mechanism matter? If you have even the remotest notion of "Rolling Your Own" Oracle Database "Recoverer", - don't! RMAN is one of a very few options that Oracle will support. Try to do almost anything else and your on your own and that's not somewhere you want to be when your Production database goes down.