John I was just thinking yesterday that paralleling coils would raise the voltage they see vs running them serial (and so increase speed), while the full circuit circuit current could increase proportionately without exceeding the current/heat limits of individual coils.
And sounds like that is what you've done.

Another thought...... maybe make a prony brake from two sticks of wood bolted together, and a hole drilled through the joint to make a shaft clamp. Measure out some convenient arm length along the sticks from the shaft, and put a spring scale at that point. Energize the motor, and check the scale reading. The torque is the distance out, times the scale reading
That will give you a torque measurement in whatever units you chose. Generally steppers are rated at holding torque, (and in the U.S. as in.oz.). You could check holding torque by having the motor energized but stationary, and moving the scale until it jumps a step -- that point would be the lmit of holding torque. This would give you some idea of the performance of your stepper, and also allow you to compare different speed hookups and microstepping effects.
If you don't have a scale, you can make one effectively from the brake itself. Put a spring on the brake to a fixed point, add a dial and pointer. Calibrate with some known weights, and then do your test of the motor.