Winding transformers can be frustrating, but not that bad for SMPS xformers. Try ding power transformers where you end up doing over 800 turns! Thats when a DIY winding jigs with counter comes in handy.
Alas, perhaps I missed something but you didn't specify the inductance, core material, or saturation current. In many projects I copied fom books when I was younger, they jsut said 'this core', and the number of windings. For the most part thats all you need, but given the vast number of core types, and geometries available these days, if you can't get the exact transformer core/bobbin, then you'll have to recalculate the number of turns (with the ratio being the same of course).
If you want some transformers to play with, farnell, ebay, and I'm sure mouser etc... will sell offline SMPS transformers in the range of 5W-200W. The smaller ones usually have a ratio anywhere between 8:1, and 15:1. I managed to get a lot of 30 from ebay for a few ££, and after measuring them, just unwound a few turns as needed to get the correct ratio/inductance. These were of course limited to power apps (not filters) but it saves a hell of a lot of time as the windings were all neat, well insulated, and professionally produced - I just modded them as needed and it took maybe a minute. Please note: most if not all SMPS power transformers are 'flyback', meaning they are basically coupled inductors rather than true transformers - the core will have a gap, and the inductance will be very low (down to 2uH for the primary windings). For ungapped cores, a few turns quickly goes up to mH. I'm guessing you'll need an ungapped core.
It all depends on your application. I'm assuming you need power conversion. And the 'figure 8' cable is, I can only assume to be what was known as 'bell wire'. Its literally two wires with their insulation glued together. This means, you can wind 8 turns with it, and you end up with two coils of 8 turns each. By connecting both these wires at each end, you have a single 8-turn coil, but with double the current capacity (two wires in parallel, same resistance = 1/2 resistance). But if you connect ehn in series, you have a total of 16 turns, with a 'tap' connection in the middle of each.
Looks to me like a high voltage application using a royer oscillator. The secondary is wound first (because you want it to pick up all of the magnetic field from both the core, and the primary) and thats 90 turns (two layers of 45). The primary is tapped, that is, two 8-turn coils, connected in series. What makes this seem complicated is that, its just easier to wind 8-turns of 'double bell wire', becuse you're winding TWO wires per turn. Then these can be conected as per the diagram.
Different wire diameters? In terms of transformer operation, it doesn't matter. However! Thinner wire has mroe resistance, so it all depends on how much current you want to pass through the primary. The secondary will carry less current, at a highervoltage, so it can be relatively thin. But as the rato is 1:11.5, your primary must carry 11.5 times more current. Too thick wire and it becomes difficult to wind, and may not fit inside the bobbin - too thin, and the efficiency of your power conversion will suffer, because the wire's ressitance will convert current, to heat.
The difference between 0.2mm and 0,25mm wire is small. I would say use just 0.25mm for both. This means the primary winding (recommended at 0.25mm) is fine for its rated current. But the secondary winding wire will be thicker, so you won't be able to 'fit' 45 turns per layer, and also each layer will be thicker. Sooo you'll have to recalculate how many turns per layer, and there for how man layers in order to get your 90 turns on the secondary. I believe they chose the wire diameter, not forcurrent capability (the secondary can be much thinner than 0.2mm) but purely for mechnical purposes - that is, 0.2mm wire would mean each layer can be exactly 45 turns, as you need 90, this gives two full layers of wire - which is neat.
Depending on where you are, it shouldn't be too difficult to get hold of enamelled copper wire. I'm a fan of Ebay shops, and in the UK i can pick up 50 metres of -.1-1.0mm for around 2USD. Some will even sell shorter lengths for a cheaper price.
With transformer winding, even without 'electrical properties' it requires a bit of basic maths to calculate the volume of the area the wire will occupy, and indeed, wire gauges/diameters are often chosen for convenience.
If you provide the measurements of the bobbin you're using, I can go off on a tangent about calculating how the turns are wound depending on the wire diameter
Blueteeth