"I want to drive the transformers individually using synced clocks (such each transformer has its own IC driver and opto-feedback) ..." For the secondaries to combine additively all the primaries would have to be driven in-phase and in-sync. Without that the secondaries will "beat" sometimes adding and sometimes cancelling each other out. Edit your question to explain further.

The schematic is not correct then, as the transformers are driven independently, not literally in parallel; please show the actual driver / controller circuit used. The output, is it AC? Is there any requirement on waveform per tap?

Any particular reason not to just wind your transformer for your desired output voltage?

I have edited given your guy's comments. Hopefully this is clearer

So you want a number of independent/isolated 50V outputs that can potentially be connected in series (and disconnected whenever needed)?

Do you need every 50 V step all the way up to 2 kV? As in 40 outputs in total?

@winny No, just up to 600V. The 2kV is the galvanic isolation rating of the transformers between the primary and secondary. Rohat - No, they are always in series

2 kV is probably the hi-pot rating. 600 V isn't sky high, but be aware that continuous use != hi-pot. Do you need ever 50 V step up to 600 V, as in 12 outputs in total?

@winny Yes, continuous use of 50V per channel totalling 600V system voltage

Have you looked at using a Cockroft-Walton ladder with 50V per stage? en.wikipedia.org/wiki/Cockcroft%E2%80%93Walton_generator

Yes, it's possible, but you need to consider edge cases as one of the middle rails overloaded, goes into OCP and if you have any load across your 600 V total rail to ground, that will try to force current though the output of the middle one in OCP. Similar issue as batteries in series and one gets depleted before the others and it will see negative voltage. You need at the very least a diode across each output to bypass that current.

@winny Could you please add an answer with the risks/issues as you have mentioned above? That would be great to hear/understand these types of risks

Would it be better to drive all the primaries from one switch (if it can take the current)? and only have feedback for the lowest flyback as they should all be balanced.

Right, now that you've edited the question, it's clear that you are not series-connecting transformers but series connecting SMPS DC outputs. That changes everything as the outputs are smoothed and there is no need for synchronisation. Please fix the title of the question - or if you do mean to series-connect the transformer secondaries then make that clear in the post.

@Transistor Yes, the DC outputs. I have edited the title

@Transistor some people use "transformer" instead of "converter".

Are the component values and part numbers representative? I find it hard to believe you need a 2.5mΩ MOSFET to switch a measly 2.5W per stage! || Your question asks "feasible" -- by what metrics shall we measure feasibility?: basic function, precision, efficiency, layout area, parts count, etc. Some more background about what requirements you have for the numerous outputs (per each and in total, current range, accuracy, example loads) would also help greatly.

@RohatKılıç, I know. I'm changing the world! d:^)

@TimWilliams Hi Tim, no the numbers are not representative. I am more interested in the topology pro/cons before I go into detailed design. However, the image is representative of what I am looking to implement. By feasible, I primarily mean runs fairly predictable and stable, safe and does not have any silly drawbacks like 10% efficient.