or perhaps Perl 6 considerations I might want to address early

That's certainly a topic I would like to hear about. We recommend people to try to avoid naive regex based parsing and lexing all the time for anything that is moderately complex, and to use a proper dedicated tool instead, like P::RD. But the whole idea of regexen being promoted to "rules" possibly gathered in grammars sheds a new light on the issue and indeed in a very exciting manner.