The injuries and damage caused by a small explosion and exposure to a 400 knot wind blast display remarkable similarities, especially if a flash fire also erupts. The cellulose coating on playing cards can give positive readings for explosives as evidenced by the IRA bombing convictions in the 1970s which were subsequently overturned many years later. Plenty of other common aromatic chemicals can cause false detection of explosive residues, so these all have to be eliminated by minute and careful analysis before a definite statement can be made as to whether explosives of any sort were involved.
I doubt very much if a surveillance satellite could detect the heat signature of anything less than several kilos of explosive under daylight conditions. It is far more likely that the two heat flashes were the fuel tanks exploding in the air and the final impact on the ground.
Several kilos of explosive would have left obvious visible marks on the airframe and passengers, whereas sub kilo amounts would be unlikely to cause an immediate break up of the aircraft. Cargo holds and floor panels have been reinforced and blow out panels fitted since the 747 was brought down by a bomb at Lockerbie.
Some form of structural failure is still the most likely cause. Although composite components have been used in aircraft for many years, there is still no complete understanding of their fatigue life and failure mechanisms. Internal delamination of parts can go completely unnoticed under visual inspection and without subjecting them to ultrasound and other non destructive tests such faults will continue to propagate invisibly through the component until sudden failure. Such tests take a lot of time and need expert interpretation, therefore costing a lot of money to conduct.