No one here has mention it here but Dragon 2.6 was realised in December. Its is now rated above Fat Fritz 2.0 but still bellow Stockfish. I have not tried it yet.

Playing some moves in Corr Chess I notice Dragon 2.5 its not only a bit faster than Dragon 2, it even reaches high plys about as fast as Stockfish 14.

I had no problem, but the net is bigger.

I can add one last thing. In two Correspondence chess games I can explain better the differences between engine types. 

In one Correspondence Chess game Fat Fritz 2 saw a mate in 11 almost instantly while Dragon 2 did not see it proposing an other move that also would have won very easy but slower. But in one other Corr game Fat Fritz 2 was extremly slow(!) and thougt winning for me but did not see the move. Dragon 2 suggested an other move than Fat Fritz 2 and after I enterted the move Fat Fritz saw the strenth of the move evaluting as even better for my than its own suggestions.

Engines based an calculating long lines fast can spot forced mates or long tactical lines fast while other types of engines can discovered better positional moves.

Since you believe, "Long live the Nimzo-Indian," have you ever tried running lines of it through Dragon 2 vs Stockfish? White's choices for move four can lead you deep into the rabbit hole whether you play white or black.

I am not an engine expert or engine developer but from what I have read there are several approaches to create engines.

Each and every engine (and clones) has unique way to store the board interal , an unique way to suggest candiate moves and unique way to evaluate the position.

Some engines undertands that a move is good based on evaluations at large depth. Some engines understands that a move is good by evaluating the current position by more parameters  thus trusting its own assesment more and som engine does a more broad search thus impossible to get high depth. Both appaches has strengths and weaknesses, if not looking to deep best move may be missed or looking to deep horizon effect can acour and missing good replyes on the way.

The different approaches leads to different strengt,  in very tactical position I would prefere a higher depth based engine while in imbalanced or more postional position maybee an other type.

I think that is mostly the gist of the idea.