If you have GC-MS software and MS library you can use it to find the expected mass spectra of your analytes. Select two m/z values for each analyte set the MS in SIM program and that is all.
You can check also the literature for product of the derivatization, maybe somebody already selected the optimal conditions (SIM) for its detection/quantification.
To analyse a range of acids above C20:0 you do not want to use a SIM method as they will be very hard to identify. I have successfully analysed silylated fatty acids up to C34:0 using a DB-5ms column and a generic analytical programme. A 15mx2.5mmx2.5μm column (or a 30m version with a lot longer run time) is suitable and should allow the longer acids to elute. I usually use a split/splitless injector at 250-300° or programmed to follow the oven temperature; gas flow should be set up for the length of column; oven temperature programme 2 to 5 mins isothermal at 50°C (depending on how much delay you have to allow for the solvent peak to pass before starting to collect data), 10° per minute up to 340-350° (depending on the top temperature of your column), 10 mins isothermal hold.
The mass spectra of silylated fatty acids are very distinctive with a large peak at 117 (all acids) and a large peak at [M-15]+; the molecular ion is usually present but small. For C20:0 [M-15]+ is 369 and you can easily work up or down from C20:0 by adding/subtracting 14 for each extra or less carbon atom.