Actually, there's a lot of frequency reuse on MARCS, so there are primary frequencies which are alternate frequencies elsewhere and vice-versa. You could probably re-work your file to include the whole list (primaries and alternates, minus duplicates) and be just fine; the list I use has a total 227 frequencies in it with no duplicates.
Only once did I have to edit for space (more control channels than the scanner would hold), so I just edited out the alternates of some sites in the middle of nowhere until I reached a quantity the scanner was happy with.
Update: With new sites having been added I redid the list; there are now 256 control channels and alternates, which includes all of the alternates to the alternates at simulcast sites; real radios only get the primary and secondary, not the alternates to the alternates. If I trim out the alternates to the alternates, it probably comes out more like 230.