If the ridge over Texas shifts eastward, then that would likely induce a temporary turn back to the WNW before the storm resumes moving NW around the high. If a high simply builds in from the north of Rita, then the current motion would likely continue with perhaps a further east landfall compared to the first scenario. The latest GFS run suggests a NW motion more towards western LA, but then a slight shift westward as it gets closer to the coast, taking the storm in near the TX/LA border. Whether this happens, I don't know, but it is possible that Rita will take more a stair-step motion towards the coast, moving NW in the near future, then turning a little more to the WNW, before finally turning back NW before landfall.

The latest GFS also slows the system somewhat before landfall and stalls the system not too far inland. The trend for slower movement before landfall would suggest a weaker storm at landfall, but a prolonged battering of the coast and excessive rainfall. Regardless, it appears that the steering currents will become less defined with time (perhaps that is already happening), which increases the possible error in the forecast.