Ian, The reason for having many small tubes is to increase the heating surface, and also on a relatively short length with a large tube, a lot of the hot gas can pass through without getting near the surface. The boiler I had in the picket boat is roughly the same size as in the sketch, and when running at 30psi, the temperature of the exhaust gas was around 140 C which I found surprising as the temperature of the boiling water at that pressure is 134 C. So it seems as though the efficiency is pretty reasonable.
There are no cross tubes in the main flue as that would stop me using the poker burner. Another reason is that built the way shown, all the joints are accessible. With this type of burner, there a whole lot of small fan shaped flames impinging directly onto the heating surface. In order that all the slots give a flame, practically all the air has to be taken in as primary, otherwise there would not be enough getting to the furthest slots. As you can see, the end of the flue is blanked off with only a couple of small holes for the secondary air. In fact, the burner will not light if it is out of the boiler. I feel that the blowlamp type of burner you are using just blasts down the main flue without doing much heating at the input end.
I don't know about gentle heating, it's just possible to look into the flue through the secondary air holes, and when working well, all one can see is an intense purple haze.
As you can see from the photo a few posts ago in the picket boat thread, the picket boat goes very well. There it is running at about 30 psi and the engine is the 5/8" bore and stroke twin also pictured.
Kno3, the "inner dome" is a baffle surrounding the water level probe. With fast boiling, the actual water level is rather difficult to detect, and as you can see there are a couple of 1/8" holes top and bottom so things are rather calmer around the probe.