Your approach in either case is correct. If you're assuming a very highly ventilated cavity, you can just ignore the layers outside the cavity as you said, just make sure you also turn off solar radiation in addition to the parameters you mentioned.
If you want to simulate the cavity and outer layers (as you would need to if you were concerned about the outer leaf), then you're correct, you model an airspace with an air change source and then whatever layers exist beyond it.
The trick with the second method is knowing what to use for an air change rate in the cavity. This will depend on a number of factors. There is an interesting paper out of Oak Ridge National Labs in America that discusses cavity ventilation and references ventilation rates from 0.3 to 50 ach depending on the conditions. You can find the paper at:
http://www.ornl.gov/sci/roofs+walls/sta ... rs/210.pdf
It also include references to a number of other sources that may be useful.
Hope that helps!