I think the big risk here is the OP's assertion that he was considering this in pursuit of longer runtime
. This won't help him, since his runtime will be limited by the *6V* battery that he has in series with the current 12V battery.
I would say it's fine *IF* you have a way to ensure you don't over-discharge the 6V battery (and damage it).
However, the benefits are fairly small:
(A) It will not increase runtime by any substantial amount since you are still limited by the 6V battery
(B) It will increase the lifetime of the 12V batteries since they won't be discharged as far (degradation is lower on an SLA if depth-of-discharge is lower for each cycle).
(C) It will improve the Vdroop some under high loads (unlikely to be noticeable)
I would recommend against it if you are not actively monitoring the 6V battery, because you're more likely to run it very low since the 12V will not be "getting low" around the same time -- therefore the decrease in speed/torque/etc. will be less noticeable than your current setup. Running the 6V too low will at least reduce its lifetime substantially.
Now, you could also go buy another 6V, 12AH battery and wire that in parallel with the current one...
F-150 - 24-36V with homebrew 100A variable-speed motor controller, 4x 775 motors (no, it's NOT all-wheel-drive--look up 'quad damage' in the gearbox section for more info!)
Pink Rocket Princess Mustang - 15mph, 36V, 4x 775 motors, homebrew PWM controller rev 2....
Hurricane - 24V w/50A step-down controller, 775 motors