You say: "I can "believe" that transmission line effects might cause a double size voltage for one cycle or so ( ?) of the 50 Hz but would expect that the output of the generators / transformers distribution system, long enough to destroy a household appliance under even no load conditions would be no more than say, 110% of nominal."

Sorry, but you are hopelessly too optimistic. Surges on networks are far more complex than one might realise, and very hard to control on a large network. One chap I knew did some research on the problem-prone reticulation supplying a local steel producer. He found that every time they switched off an annealing oven, it caused a spike of some 14000 volts on their nominally 250 volt line... So much for 110%! I never heard what measures were taken to shut off the ovens more gracefully after that. Notice that although switching on the ovens caused voltage drops and swings, they caused nothing like such spikes.

Think of what happens if you abruptly  shut a domestic water tap. Your static water pressure might be a couple of bars, nothing too exciting, but get the water flowing and if you stop it abruptly you get loud water hammering, and sometimes actual damage to plumbing.Such effects have been known to burst piping.

Now, electrical reticulations behave in very similar ways. Suddenly break a circuit and even a small battery that could not cause a spark to jump no matter how suddenly you short-circuited it, will cause a visible spark, a spark that would require tens or hundreds of volts to create it. What you are looking at is analogous to the water-hammer effect. You stopped the electric current, and the energy that had gone into building up tha current had to go somewhere!

Jon is precisely right. Apart from plain "thermal" (resistive heating) loads, almost everything else (motors, fluorescents etc.) works by inducing magnetism through coils. All those devices are designed to match their impedance to the 50Hz AC frequency, and the magnetic fields absorb power just like an oscillating spring. Breaking the circuit dumps all the energy stored in the circuit at once. All that energy goes back into the source cables in a millisecond. And of course, having your fridge and washing machine turn off makes a tiny spike that was designed for. Losing supply to maybe 500 homes is cumulative - everybody's spike adds their voltage back into the distribution network simultaneously. Major distribution switches are designed to operate at the zero volts point in the AC cycle to minimise this effect for planned switching, but cannot do much about sudden circuit breaks.

This is actually how the coil in your car makes 20,000 volts from 12 volts for the spark plugs: the coil magnetises from increasing a current relatively slowly, but the circuit is broken (by points or electronically) abruptly, and the high rate of change in the magnetic field as it decays induces a high voltage.

The fires and further damage are not caused directly by the pulse (although it can blow memory chips and other vulnerable parts). What happens in a domestic unit such as a TV is that the initial over-voltage flashes over to earth, damaging the component insulation by leaving a carbonised track. Normal mains voltage then leaks through the insulation at this point, causing local heating and cumulative damage until something gets too hot to live.

The pont is, that I am suggesting  or querying, there cannot be a POWER  surge. There can be a voltage surge but only with the total energy stored in the inductance of the distribition lines by the carrent interrupted. I understand that this voltage surge can destroy insulation. 

Surely, the water analogy is not valid. The "pressure" of the water is reversing 100 times a second.