You can certainly make the 400rwhp number with a 357W - it just takes 500fwhp to do it, given the auto tranny. You will find that a 450fwhp number is pretty straight-forward to achieve based on any number of similar parts combos. That is, it doesn't take exotic stuff to pull it off. 500fwhp is tougher and requires more thought, or a different kind of build. It certainly is more expensive. The key to the power will be the header/heads/cam/intake combo. The power number will probably come at >6000RPM, so you have to plan accordingly, which is why more money will likely be spent, especially if the build is in the hands of an amateur builder (like me). A professional engine builder is able to do more custom work (porting, better cam selection, for example), that's how they make their living - by being better at it than we amateurs are and not having to pay someone else to do it. We generally have to throw money at it to achieve the same thing. That's not always the case - there will be exceptions - but I think it is the rule.
As for 393 vs. 408: BottleFed70 answered for me pretty well, but my answer really depends on the purpose for the engine. If you drive the car <5000 miles a year and aren't really interested in an engine capable of running 50-100K miles, then the answer is that there is no substitute for cubic inches. My personal goals are to have an engine that is just as strong running at 25K, 50K, etc. miles as when it is fresh. I mean, that's the plan. So I try to keep the rod ratio at 1.65 or better and the piston compression height 1.25" or more. Now, I break those rules, too, but that's the general idea. My current 331ci motor has a rod ration of 1.66 (good), but a c-h of just 1.175. So, I made a compromise. I was not willing, however, to compromise to 1.09" c-h or 1.59 rod ratio, which is what you typically get with a 347. Also, I consider readily available parts and their costs. 393 kits with a long rod - I distinguish this from the 302 rod based kits which I would not use - are readily available, have a good c-h and a pretty good rod ratio of 1.61 or better, depending on the rod (6.2" or more). I would build this engine so long as I was building a <6500RPM motor. If I was building for higher RPMs, I'd look at the 383ci. This is all just my opinion, so take it for what it's worth.