Low loss header feed and returns

[The Viking]

OK,
Regardless of theories of how it might work, or not as the case might be, the proof of the pudding is in the (h)eating.

Correctly designed and installed there should be no temperature differential between the primary and secondary flows.

[nippyjock]

See what you did there, I am a lover of puns myself, not so keen on non-specialised specialists, (aka my installers).

[nippyjock]

MikeHolm, You stated "the two feeds out to the load need to be quite close together and they need non return valves on them. ALSO, very important is that the flow from the two pumps should each match their respective heat loads and together needs to roughly match the total max heat load which, theoretically, should be less than the flow rate on the primary side. If you can slow down those secondary loops pumps, you will fix most of the problem."
I agree these are crucial to system optimum performance and efficiency, however, would you consider that these issues are secondary, i.e. not the root cause of the problem, and the biggest impact on performance and efficiency is the primary flow and return being crossed over?
To me this appears to be the case, my 55C flow temperature is getting nowhere near my secondary heating circuits, it is being cooled immediately by the secondary returns, in reality my secondary feed circuits are typically 32C when the heating is first switched on.