The reason the mic is at such a close proximity was established by the poster - the scale of his sensor did not allow for any readings at a more reasonable distance. Consequently, the "arbitrary positioning" of his sensor was the first source of affected information - whereas a distance of, say, 3ft would have negated the location of internal parts and air circulation as a factor, such a close location makes the significance of sounds from whirring drives, spinning fans, air blowing directly onto the sensor etc a problem. For example, in the case of the 360, the sensor has half the console between it and the disc drive, whereas the sensor is right next to the casing where the PS3 disc drive is. At such short distances, 1 or 2 cm this way or that could be quite significant as well. The sensor seems to be below the majority of the Wii for a start.
With the different physical construction of each console, it is impossible to find a consistent position for all three consoles. The fact the sensor itself is resting on the shelf is also an issue - I doubt any of the consoles would have found the resonant frequency of their respective shelves, but if I were performing such an experiment I would simply have put a kitchen/bathroom sponge underneath the sensor to dampen any effects from physical contact with the shelf.
The solution, as I said, is to use a better sensor from a greater range to nullify any issues with component locations, and put it on a dampening material.
The shelving is also a factor in terms of ideal environment for the consoles - whereas the 360 is on a larger wooden shelf with an open back and a decent distance between the shelving and the back wall, the PS3 and Wii are both in far more compact spaces on what doesn't appear to be wood (looks like plastic to me - I doubt it's metal because besides the impracticalities, that'd surely be a safety hazard!). Putting the units in such different environments leads not only to different levels of circulation for all three, but also makes the Wii and PS3 readings more susceptible to any sound reflections from its' surroundings (that may not be a problem for a relatively insensitive sensor, but it's something I'd be wary of anyway as a one-time lab technician and Physics graduate who has to right lots of silly reports and be as anal as possible!).
Solution: Put the consoles in the same environment to avoid circulation issues, ideally on a wooden or glass platform (carpet would retain heat much like a sweater does) and a couple of feet away from any walls to avoid reverberation.
Lastly, a graph of sound variation against time with a log of events at each time would be nice!
Ultimately, I'd just say - go by your ears. It's pretty much as scientific as this experiment, and more relevant because the conditions in which you're listening to your console(s) are the conditions relevant to you. The general consensus is that X360 is louder than the PS3, which is slightly louder than the Wii (though I'm sure people will dispute that in accordance with their differing gaming environments). As people mentioned, "louder" does not mean "more noticeable either" - I can't sleep if I am wearing a ticking watch or there is one within 3 feet of me, but I sleep like a log even if I have a fan on.
