I agree with all that is said about RNFSAIC. They have a superb reputation, world wide. As a benchmark, they are ideal, so the suggestion of a Tri-Service organisation based on this model has merit.
But I do think it wrong to suggest they can and do investigate wherever they like to uncover the underlying causes. I think, like BoIs / SIs, they are influenced too much from above, and boundaries set which they shall not cross. In the RAF domain, the classic example is the Mull BoI, which barely touched airworthiness and, indeed, MoD have always admitted it was not in their remit. In other words, they were told not to go there, as it exposed too many embarrassing truths. Independence from higher influence is needed.
If one was to read the BoI and RNFSAIC reports on, for example, the Sea King ASaC collision, written 2 years apart, there are a ludicrous number of contradictions between them (and within). One, the interpretation of a tape recording, was misinterpreted so badly by the BoI (for entirely understandable reasons) that it clearly influenced their recommendations and the tone of the report.
In such cases, would it not be proper for either the BoI to reconvene, even if only to issue a clarifying statement or acknowledgment; or is it for the Convening Authority to reconcile the differences and issue a statement? After all, their common goal is to prevent recurrence. If you don’t get to the root cause(s), then this basic pillar of Safety Management cannot be satisfied. But, there again, in many cases the Convening Authority / Reviewing Officers have too much to hide. Which brings us back to independence.