Thanks everyone, and good spot with the lights being screwed in place Chris
I'm not sure I will do any moving closer to the lights though. The thermometer in the cupboard reads 26-30C which I'm happy with, so heat is not an issue.
And I'm not convinced by the 'light falls off as a square of distance' argument in this case. I shall now go into a long explanation of why & might be wrong
So we've got a big brown propagator, two bulbs, the 1st horizontal is the LED panel I made (which is mirrored perspex), and the middle horizontal line is an elevated position for the propagator.
Now, light from a point source falls off as 1/(r*r) because the surface of a sphere is 4*pi*r*r, as the distance is bigger the surface of the sphere expands etc. But the important thing (I think) is how much of the sphere's surface my propagator intersects, and the fact that the cfl tube is not a point light source.
The light I get with the propagator in the lower place looks like it gets about 50%-65% of the light the middle horizontal gets. So far, moving closer is good!
But if I add something reflective down the sides, then the 50% of total light lost there is back in play... and should lead to a more-than-doubling of light on the plants?
Light doesn't get absorbed by air, the light from the bulbs is just more spread out, so adding something to bounce it back again should be more effective than moving closer to the source?
The cupboard it's in is painted with highly shiny white gloss (doesn't come across in the photo), but something more reflective might help (I had foil last year, but it's a bit of a pain)
So, the long and the short is I'm gonna struggle on with my 'too far from the bulbs' approach and see how the plants do
Of course none of this means I'm right, and thank you all for your help and advice!!!