Ghostflyer and Stormcloud - sorry if I didn't make myself very clear (I was trying to keep it simple for Airbubba!).
The company policy at BA is "any engine failure/fire = land at nearest suitable airport". The company line on that is "look at the nearest airport and go there if it is suitable". However, the crew and aeroplane should (if possible) be fully prepared so as to avoid a rushed approach.
So, in the scenarios Ghostflyer proposes, scenario number one is the ideal but you go for number two if the fire doesn't go out. In Ghostflyers more detailed scenarios, I tend to agree that the second case sounds better - but you have to balance the desire to land an aeroplane that has had a fire ASAP with the need to do it safely somewhere. This is for the crew and the crew alone to decide, in my opinion, bearing in mind the company policy of landing ASAP (which I think is VERY sound advice!).
For Stormcloud - a gravel strip would not normally be regarded as "suitable" - but we were, in fact, discussing "putting it down on the nearest piece of tarmac", not a gravel strip if tarmac is available/reachable/useable. I would further clarify that as the nearest "suitable" piece of tarmac.
The grey area comes with "suitable" and of course that is a question that can only really be answered by the crew in the aeroplane on the day. Weather, runway direction/length, approach aids, crew familiarity, political considerations (do you want to use Baghdad International?), ground services and facilities will all enter the equation and no two scenarios will ever be the same.
I myself have landed an aeroplane that had a double engine failure (we started with 4!) and a big fuel leak at an airport other than the nearest. However, the airport we chose was (in our opinion as the flightdeck crew) the most suitable because: we had just started the descent for that airport, we already knew the weather and had briefed the approach, it was home so we knew the airport inside out, the emergency crews knew our aeroplane inside out (as did the engineers on the ground) and a diversion to the nearest airport would actually have taken longer because of the need to re-plan and we could not spare the fuel.
If we had just reached top of climb and had not already prepared for the approach, then the choice could well have been different.
As yet we do not know what decision making process the crew went through (and probably never will) but I very much doubt that they had any priority higher in their minds than flight safety.