My ad hoc rule is to have about 1 baseline parameter per 150 s of imaging time, with a minimum of polort=2. (I'm assuming a normal TR in the range 1-4 s.) For 14 minutes = 840 s, you might try polort=6 based on this rule.
However, I would recommend strongly that if you use such a large polort, you also output the time series fit dataset (the "-iresp" option) and see how well the resulting model works. In fact, I would generally recommend this step for any modeling analysis: see how well the fitted model looks agains the real data. This is easy in AFNI, using the "Double Plot" option in the graphing window and the "Dataset #N" plugin.