I will put up a small argument on the 4.4V thing. Reciever dropout (loss of signal from the transmitter) becomes a real factor at 4.6V. At 4.8V depending on how the plane is flown is getting close to the time I charge. Mild standard non-digital servos do not draw much current when moved one at a time. Now start moving them all at once and see how much voltage drop you get. Even a momentary drop to 4.4V for just a few milliseconds is enough for the receiver to drop out. Five JR537 Standard servos moving all at once can drop the voltage by .3 - .5 volts for several milliseconds and that's assuming there is absolutely no binding in the control rod or surface. Any bind what so ever and the amount of voltage required goes up not to mention the air loads on the surface creating a need for more voltage if the model is flying.

Once the receiver drops out (loses signal) from too low a voltage there may not be enough voltage left to get control back since now you are in panic mode and moving everything to try to get control back. Or you go into the dropout/get-it-back/dropout round robin and once again you have very little to no control.

I will not even get into what digital servos will do.

To me, it's not worth the risk going below 4.8V and with a decent field charge, you can be charged in an hour or less.