I'm not sure I really know enough about detailed locomotive design to argue my point firmly. I agree that if you've stuffed up your 2-2-2 design with too much weight on the leading axle than a 4-2-2 conversion is an obvious fix (pace William Dean) and the same is presumably true of 2-4-0s. But I'm not really thinking of conversions, even paper conversions, but of clean sheet designs. Going back to the 6 wheel types, the similarity in general layout was such that the GWR could convert redundant 2-2-2s into 0-6-0s by pointing the cylinders at an axle about a foot lower. I submit that the major differences between 2-4-0 and 0-4-2 was that the centre of gravity needed to be a little further forward to get the weight distribution right. Cylinders, motion, valve events, draughting, all these things were fundamentally the same, and that if a locomotive superintendent had said to his staff design an 0-4-2 based on our 2-4-0, both with rigid six wheeled chassis, they wouldn't have found the task very onerous. The reverse is arguably a little more difficult, since there is a certain amount of potential complication with firebox, ashpan and rear axle, but, I submit, not much more challenging. I suggest that there's no point in building a 4-4-0 (or 2-4-2 or 4-2-2) if a 2-4-0 will do the job, so features like a larger boiler/cylinders which would make a 6-wheeler front heavy must be the reason for the design. So let's put ourselves in two design offices, one with good 2-4-0s and one with good 0-4-2s (and both with good 0-6-0s). The superintendent comes in and says he wants a 4-4-0. Both teams have to discard the leading axle of their type, and rearrange cylinders and front end frame design, then deal with side control and all the aspects of bogie design. That, I suggest, is the real meat of the job. They can probably retain the same design principles in cylinders, valves, motion etc. The 0-4-2 team have to replace their trailing wheel with a driving axle, but as they are already familiar with such on an 0-6-0 will it really cause any issues for them? So that while on the one hand, yes, you can successfully argue that the passenger express engine without guiding wheels was a dead end and wasn't going to make the transition to 8 wheeled types (although that doesn't seem to have marred 0-4-4Ts) my suggestion is that in practice the length of the dead end siding was minimal and there were no issues in returning to the main line of steam locomotive development.