First of all, pinto itself it's just a tool to handle cpan-like repositories, it does nothing with HOW source code gets built and becomes distribution archives. So pjam is upper layer to integrate sources code changes (SCM) and pinto, which in turn produce builds (distribution archives). Right now it's only about subversion, but it can be generalized on other SCMs.

Secondly, pjam is helpful abstraction to organize source code into structural projects to simplify distribution processes.