Now to construct MY model

The two entity model repetitive prisoner's dillemma has the major fault of not counting the third party terrorist organisations. Now since these are primarily funded by outside countries completely outside of the control of Israel and PA (because as you have said he who pays the piper calls the tune) they must be considered as a third entity with fairly different scoring alternatives. Where as the PA would have benefitted substantially from the Oslo accord(if it worked), the third party terrorist organisations would have lost influence, especially if they stopped attacking Israel. A win for these third party terrorist organisations is essentially the "destruction" or making illegitimate of Israel. I think this can be modelled in a three dimensional matrix, where the PA can choose to attack or negotiate, the third party terrorists can choose to attack or not to attack, Israel can choose whether to attack or negotiate. Note that even when the PA was heavily armed, it was (EXTREMELY) doubtful whether they would wage war against the third party terrorist organisations. If it came to a choice (at a police/soldier level) of shooting Israelis or shooting their "own" they would shoot the Israelis. I'm not just speculating about that - there was a couple of examples during the Oslo peace process of it. This three d game theory model effectively gives the third party terrorist organisations complete veto over any peace deal that involves them stopping their attacks. The reason that they would temporarily stop is to loosen some of the security measures and/or find new holes. If there is anyone in the University of New England interested in game theory, this is the one to research. While there are groups like Hizbullah (which is Iran-backed) that are backed by countries that haven't made deals with Israel, no deal that requires the PA to stop attacks are possible. Maybe if this model was played out, and compared to what has happened over the last few generations, it would fit perfectly, whereas the two entity model (or the three entity model with USA as the third party) fails to predict the non-signing of Oslo, for instance.