You're of course right about growth curves being logistic functions with a plateau at the end.

The question is however, where the plateau lies. Perhaps it's so far off that for practical purposes it doesn't really matter that there is an upper bound at all.