Just a word of caution when twisting extention cables. Twisting can put a lot of strain on the end of the wire where it is crimped or soldered to the pin. I was involved in twisting thousands of cable from 1968-1983. We used a tool that clamped down on each wire just where it exited the connector shell so the strain was not applied to the connection to the pin or socket.

In my opinion a lot of the problems associated with long leads is caused by the fast rise time of the pulse produced by the decoder IC. We used a .001uf cap from each signal lead to ground in our receiver to roll off the rise time a bit on the radios we built. I know the .001 cap reduced the trash on the signal going to the servo. I have run servos over cables as long as 150 ft.

Twisting the leads adds capacity between the signal lead and ground. One disavantage to twisted cable is the larger amount of resistance for a given length of cable which will cause a larger voltage drop to the servo.

In a model I am building I plan to use an 18 gage twisted power buss from the batteries in the nose to the servos in the tail which are about 80 in. apart. I will also run a 26 gage ground wire twisted with the 26 gage signal lead for each servo.