FF4 without Reflector provides UHF Raw Gain = 9.7 to 12.6 dBi:
UHF Free-Form 4-Bay Bowtie - NO Refl
FF4 with Optimized Double Angle Reflector provides UHF Raw Gain = 14.0 to 15.5 dBi:
UHF Free-Form 4-Bay + DblAnglRefl - OPT
So Optimized Double Angle Reflector provided an improvement of about 4.3 to 2.9 dB.
Horizontally Stacked FF4's (aka H-FF8) provides UHF Raw Gain = 16.9 to 20.1 to 19.8 dB [35-in]
or UHF Raw Gain = 17.3 to 19.6 to 19.4 dBi [51-in Stacking Distance between Center of each 4-Bay]
Horiz Stack 2xFF4 + DAR + HHH
and Vertically Stacked FF4's (aka V-FF8) provides UHF Raw Gain = 17.5 to 19.9 dBi:
Vert-Stack FF4 + DoubleAngleRefl.
So Horizontally Stacking FF4's improves Gain by about 3.0-4.5 dB.
And Vertically Stacking FF4's improves Gain by about 3.5-4.4 dB.
===================================================
FF6 without Reflector provides UHF Raw Gain = 11.0 to 14.5 dBi:
UHF Vertical 6-Bay Bowtie - NO Refl
So FF6 WITH Double Angle Reflector "probably" increases Gain by ABOUT 3.5 dB,
hence we should expect UHF Raw Gain ~ 14.5 to 18.0 dBi (more or less).
Horizontal or Vertical Stacking FF6's "probably" adds ABOUT 4 dB more: UHF Raw Gain ~ 18.5 to 22.0 dBi.
===================================================
So FF6's MAY only provide an improvement of perhaps 1-2 dB versus FF4's....but if you insist....
Following the established patterns, it's tedious but fairly straight-forward to add more HEIGHT to the Variable Double Angle Reflector (V-DAR) Model so that it extends 6-in beyond the FF6 (preferably...could shrink to 4-in if necessary). Perhaps you want to start with the non-Stacked FF6 and add the DAR....it will end up being essentially the SAME in the Stacked version(s)....which is more complicated by incorporating the Stacking Offset....esp. in the Horizontal Stack version.
If YOU modify the STACKED FF4 DAR 4nec2 File, by copy/paste of two FF6's to replace FF4's and extend HEIGHT of the two DARs, I'll run it on one of my computers...AFTER the current heat wave breaks in a few days....I don't have A/C. The Optimization Runs will take AT LEAST a WEEK....for EACH TRIAL...and it usuallly takes at least a few runs to get it right....