1. I would suspect that the switch to ITBS is to align with the CogAT, as they are co-normed, unlike MAP and OLSAT, or MAP and any other ability measure. MAP does have the advantage of adaptive testing, and thus hypothetical access to items through the upper reaches of fifth grade (for a third grader testing on the 2-5 version), but Iowas have the advantage--which is not insignificant--of co-norming.
2. The norms for CogAT/Iowas are about 3-4 years newer than those for OLSAT/SAT-10. As above, co-normed instruments for ability and achievement are generally preferred, and these are the co-normed combinations for CogAT and OLSAT. If the district doesn't want the SAT/10s, then the OLSAT doesn't allow for any other co-normed situation. I tend to favor the level of ongoing research feeding into the CogAT, also.
3. Yes. Though of course, all group-administered instruments have their flaws when assessing extremely low incidence outliers (such as 2e or PG).
