Not necessarily, at the moment there is a database with the results of the STRs for the 13 (well known) loci that are tested. So if you have access to the database you can make a match for a specific person and fake evidence against them.
But, if you have a second group of 13 loci that are kept secret, and the records for these STRs are NEVER recorded, then anytime evidence needs to be checked, the results from both groups of STRs can be checked. That information would not be kept, so it would be impossible to fake it from a database. Why even limit it to 13 loci? Why not have a list of hundreds of STRs that can be checked to corroborate evidence?