Without stirring a debate as to the practicality of S-Bus in its current configuration perhaps an analogy might help. S-Bus sends the data for all servos down a single conductor much like a freight train traveling down a track. If you want to get to the forth car (channel) in the train you have to break the train apart to get that car. Hence a switch yard. In a switch yard as the train is separated each car (channel) is sent down its correct path. Our current and familiar non S-Bus radio systems do the switching for us and each servos information (freight car) is sent to only the correct servo. On the other hand each S-Bus servo has to have the circuitry built in to pick off its information (like the switch yard). Slightly more complex than a standard servo (convert to more expensive). If you want a standard servo to work on S-Bus than you have to provide this circuitry to get your specific freight car (channel) of information off the track, called a decoder by Futaba. So an S-Bus servo works directly off the S-Bus line where as a non S-Bus servo requires a decoder to fuction on S-Bus.
Lets look at a normal wing. You probably have at least four servos two aileron and two flap. Futaba (Hobbico) in the USA sells a 3 channel decoder (Robbe in the rest of the world sells a four channel decoder). So in concept you would take 2 each 3 channel decoders (USA) and plug in/program them for your four servos (No Y harness here). Then you would plug the two decoders inputs into a terminal block in the wing. Then you could take one extension from the wing and plug that into the S-Bus of the fuselage and away you go very convenient and simple. If you had S-Bus servos in the wing you would take the four leads from the servos plug them into a terminal block in the wing then have one extension to connect to the fuselage. Again I do not wish to stir debate but the issue here is lack of redundancy and possible lack of sufficient power. If you have four servos in you wing and you plugged all four separately into your receiver theoretically you could loose one aileron connection and possibly retain control. Then the issue of power, a standard HD servo lead is 22 gauge and commonly we think of the servo connectors (the real weak link) as being able to handle about 4 amps when new and of high quality. We generally install servos with torque capability of a larger amount translate this to more current draw. Stall current of some of the high torque servos can be at the limit of what our little servo connectors can handle. Now multiply that current by 4 (your wing servos) and try to pull that through one extension (two connectors) and you begin to see the issue. a Battery could be installed in the wing or other measures taken to improve power delivery and redundancy. But then the system is no longer simple is it?
Futaba is to be commended for its forward thinking. It is up to us as users to incorporate or not based on our needs/uses. I own Futaba radios including the 14MZ and the 18MZ and use some of the S-bus on both a Nano and a Flash. S-Bus is not for everyone or every application. It is not a panacea. Power boxes are nice and have their purpose, remember there is a large component count increase in an added system like a power box, more components more chance of failure.
Gosh, hope this helps more than it confuses!
John