My point wasn't that picoLisp was in any way stopping you from using a separate database, it was that if you go the route of using the built-in abstractions listed on that page you won't find any out-of-the-box interoperability with any existing SQL database. You'll have to roll your own.
That's too bad because I think using the picoLisp database as a cache backed by a SQL database makes a lot of sense.
