Forget Ti valves in a SOHC head. Side thrusts cause them to seize in the guides.
Now I'm confused. Are we talking 350 4 or twin or a 500/550 or all of the above?
By all means red everything there is to red and then start with the basics. For the motor you have, what are the max safe RPMS based on mean piston speed. A 350 twin has a long stroke crank - for its size and the 550 isn't a high rev machine either. If you want revs, you start with a short stroke crank and then you work out what bore size works with that stroke to get the engine size you want. With say 1.5:1 bore:stroke on a 500, that means you have to use 4 valves and all of a sudden you realize that Honda and Yamaha et al already did that.
If this was 1970, when I was an engineering student and designed a 4 cylinder 16 valve DOHC 500 street bike in a chrome moly frame, it would make sense. In 3014 - not so much.
So then you ask yourself what the objective is. Is it to develop a given motor to see what works and what doesn't within some reasonable cost boundaries, or is it to jump ahead to try a combination of parts that you don't know if they work?
Let's say you were to define the project along the lines of: To develop a 1970 era motorcycle engine to maximize power within reasonable cost constraints. That way you develop a series of improvements to test on a dyno that can be published. That will not get you through a PhD thesis review but for an undergrad project it might work.
The best place for design ideas is the SAE.