This has been discussed several times but here is what I know so far.
I've found no sources that recommend another needleposition than 3rd, except... an American Honda booklet that says 4th from top. In the same settingtable in that booklet we read a recommended airscrew position of two turns out. Strange, the rest of the world has 1 turn out +/- 1/8 and the airscrews and pilot jets are the same (I suppose you have crossdrilled airscrews). It was reason for me to ask what Americans mean by one turn: is it 360o or 180o rotation? People responded and said one turn means 360o rotation.
Btw, Bryanj (UK) and Dave (Australia) might shine an extra light here. If that booklet is right for USA, it looks like USA desired a very lean idle and compensated this somewhat with rising the needle one notch. The question is: when? That booklet was edited in 1977 and since I have found no other sources before 1977 recommending two turns out and 4th, my suspicion is that Honda America had to deal with demands on emission at idle. I have no idea if the settings in that booklet describe what always has been the setting on American 500s or what should be the setting from then (1977) on. People can chime in what they've found in their 500s.
It happens that my 500 has the needle in 4th position (like you have now). Summer 2012, out of curiosity I've tested with the airscrews two turns out and the bike accelerated poorly. The 500 carbs are simple, have no accelerationjets and need a rich idle for acceleration. You can test various positions, but I doubt you will have a good accelerating bike with the airscrews two turns out. Easy to experiment and please report back. So standard worldwide is 3rd (middle) notch. For our Honda's we always count from the top and not from the tip, btw.
My 500 performs best with standard positions as Honda recommended them: one turn out and needle in middle position. It's OK with needle in 4th now, but a bit rich.
Then there is a difference in main jets: #78 for countries like Germany, Holland, Austria, Switzerland and Italy and #100 for the rest of the world. Quality of gasoline may play a role here, since there were no demands on emission in those years. Gasoline is not everywhere the same in Europe. This has to do with that engines (highly tuned) are stressed differently because of infrastructure, legislation and 'terrain'. The Germans certainly demand most of their engines. So even neighbouring countries like Holland, Belgium and Germany have different gasolines (Holland is a flat country). I've once heard from a Suzuki importer that this difference in gasoline drove them nuts in Japan.
Strangely enough changing the #78 main jets for #100 and removing the airboxcover on the "#78 main jet" models does not bring anything but extra intake noise. This seems counterintuitive. The duct cover over the air cleaner increases fueldelivery by the idle system however, so you can have a smaller main jet. Theoretically the 649A and the 627B carbs could differ in venturis, but I doubt this is the case.
It's my experience jets in my carbs work very much in a mix and not as schematic as put in graphs for carbs in general.
