Unfortunately, your formulations of the definition of strongly inaccessible are not quite correct. The correct definition is that κ is *strongly inaccessible* (also known as just plain *inaccessible*), if κ is uncountable, and whenever β < κ, then the power set of β also has size less than κ, and also κ is regular, which means that it is not the supremum of any set of size less than κ. 

This is the not the same as asserting that V<sub>&kappa;</sub> being a model of ZFC. (Although, to be sure, this false assertion has appeared surprisingly often in print and I have even heard a famous proof theorist make this assertion to a very large audience of hundreds of logicians.) The reason is that if &kappa; is strongly inaccessible, then there will always be many &gamma; &lt; &kappa; for which V<sub>&gamma;</sub> is elementary in V<sub>&kappa;</sub>, and so these also will  be models of ZFC. It is an easy exercise to show that the least &gamma; for which V<sub>&gamma;</sub> is a model of ZFC has cofinality &omega;, and so is definitely not inaccessible. 

This is also not the same as saying that &kappa; is regular and not the size of a power set of a smaller set. The reason is that if, say, CH failed, then &omega;<sub>1</sub> would not be the size of the power set of any set (since 2<sup>&omega;</sup> would be too large) and it is also regular. But we never intend that &omega;<sub>1</sub> is an inaccessible cardinal. 

Your remark that V<sub>&kappa;</sub> is close under pairs when &kappa; is inaccessible actually doesn't need anything more than &kappa; being a limit ordinal. If x and y are sets in any V<sub>&alpha;</sub>, then the pair (x,y) appears just a few steps later (and actually, one can use flat pairing function that do not increase rank at all), and so every V<sub>&lambda;</sub> is closed under pairing for any limit ordinal &lambda;. 

Finally, yes, you can apply such functions to themselves, and this idea is used quite often when we have elementary embeddings defined on models of ZFC. For example, if j:V to M, then j(j) is a function defined on M, into some structure j(M), which will be the union of j(V<sub>&alpha;</sub><sup>M</sup>).