Plenty of holes in that Chris Goodfellow scenario too mind:
1) After overflying Langakwi radar evidence shows manouvers inputted by someone on board turning sharply at VAMPI and GIVAL waypoints, past Langakwi, where that theory relies on the pilot being incapacitated by that point.
2) The Inmarsat pings recieved at 8:11am put the plane in one of two air corridors, neither of which correspond with the heading from the last known location to Langakwi, suggesting after overflying, human input MUST have occurred.
3) A fire that is able to fulfil the following 4 criteria is implausible bordering on impossible.
a) Time itself to perfection shutting down transponders/comms systems startlingly close to last radar contact point
b) Strong enough to do all damage specific enough to take out all communications systems perfectly in one go, but not touch a single censor, wire, navigation tool or flight control system
c) Bad enough to kill the crew and passengers but not hamper the airframe in anyway
d) Put itself out incapacitating the crew, allowing the plane to fly on itself for another 6 or 7 hours
Sadly it's a lot easier to debunk other theories rather than find one without any holes in it. Yet to see a single theory that doesn't have flaws. In the words of Sherlock Holmes "Once you have discounted the impossible, then whatever remains, however improbable, must be the truth." Therein we will find the answer.