Your build does not follow the basic geometry which is recommended. That's the first thing I would fix. See page 28 of "The Book" by Evans. The basic geometry is 1:2:4, or 1:2:3, depending how you wish to measure it.
FeedTube:BurnChamber:FireRiser
The Burn Chamber is to be twice as long as the height of the
Feed Tube. The height of the Fire Riser, you can measure either against the Feed Tube or the Burn Chamber. It's minimum recommended heights are either three times as tall as the feed tube, or twice as tall as the Burn Chamber.
Looking at your pictures I'd say your Burn Chamber is way too long. Most people have a horizontal length of three or four bricks on edge, where you have what, ten? And in the picture there is no way that Fire Riser is more than twice as tall as that Burn Chamber is long. The pictures did not show the finished height of the Feed Tube.
What are your measurements?
I don't think the upper barrel gap is your biggest problem. It may or may not be ideal, but a number of people report that increasing that height changes where the temperature is at is maximum in the barrel. That is not at all the same thing as saying it effects the total output of heat from the Fire Riser into the barrel. Personally, I don't see how the height of the gap at the top of the barrel can effect heat output, provided it is at least tall enough to permit air to flow unrestricted into the barrel (that's where the minimum height recommendations come from, as I understand it, at the ball park of 1.2 to 2-inches; at 2.54 cm per inch, that's roughly 3.8 to 5.1 cm, right?)
So there is a minimum height. Peter vdB has built double barrel systems in his batch-box experiments, and they have very large top gaps (measured in feet!) and they produce a lot of heat. My 6-inch system has about a 3-inch gap if I recall correctly, and it draws very well and get plenty hot. I've had it set up if I recall correctly even with a 4-inch gap at the top of the barrel, and it ran plenty hot. I personally am leaning toward a gap on the order of 2 or 3 inches between the top of the barrel and the top of the Fire Riser as my target when designing a J-style rocket heater. And 8 cm is just a small amount more than three inches, so I don't think that's your problem, in fact, I expect that part is just fine.
And insulate the fire box, it is most important that the Burn Chamber and the Fire Riser be insulated. The Feed Tube is less critical in that regard.
When you give your measurements, you ought to include the CSA of the critical elements: Feed Tube, Burn Chamber, Fire Riser, the ducting, and the amount of space provided as the transition between the barrel and the ducting (the manifold, in other words). The Wisner's have observed that the manifold is often made too small. And Peter has observed it needs to be much larger than many of us expect. Something like 4 times the CSA, if I recall correctly, but Peter is the guy to ask about that, as he's given more detail on how to calculate that than anyone else I have read.