Time for my 2 cents.

Before I start analyzing every aspect of this thread, I thought it would be fair to give a rough background of myself. I have been in the JJ2 hacking scene for many years now, which of course counts for nothing. But I am also a proud owner and homebrewer of the Nintendo DS.

The Idea
The idea of "porting", if you can call it that, Jazz Jackrabbit 2 to the NDS is not a new one. Of course, all of us here (albeit sounding very harsh in this thread) would really love to see something like this work. Unfortunately, many have tried and failed... Ok, maybe not. Unless ahat was already working on such a port. So let's further break down this analysis.

Hardware
Honestly, there's nowhere I'd rather see a JJ2 ported to than the Nintendo DS. And yet, when this is said and done, the DS is also one of the harder platforms to program on, mainly because of it's extremely limited 4MB ram. Depending on the design and implementation of the port itself, the CPU and screen size should also be taken into account.

Implementation
It is difficult to see how much thought you have put into this. I am by no means a game expert, nor have I actually made any full-fledged games. The tiles for a start; 32x32 would be pretty cramped I agree, yet 16x16 will not do any justice to the graphics. Also, the stylus controls sound very gimmicky. I'm not really sure how this will work out, since one hand will have to control a lot of buttons. A Metroid Prime Hunters style control might or might not work. Then again, don't take my word for this; I'm really bad at game design. But another thing you should consider is how much this port will be able to handle. Remember that memory is very limited. Level sizes will have to be kept small, and events (JCS events) will have to be kept few. Going in and out of levels, which many DS games do in order to load the next level, or next portion of a level, may not work in a multiplayer context.

Resources
Again, I'm not trying to completely destroy this project, but I have to wonder how much effort you have spent or are willing to spend in this. Is it going to be a one-man programming team? You did start this thread requesting for help, but the help required (sprites, sounds, levels) are nothing more than trivial aspects to the entire game creation process. At the very least, will you be designing and programming the entire thing yourself?

Programming
Finally, the section in which I'd consider myself most experienced in. This section alone can be broken into multiple subsections, but I'll have to start somewhere. Firstly, the engine. Why would the engine be written in java first, only to have it later converted to C++? Unless it was a generic engine, but this will not work very well. Not to say that JJ2 is an extremely complicated game, but the engine should be made to revolve around the game, rather than taking some engine and customising it to meet the game specs. In other words, it should be programmed directly in C++, and probably after the design stage is finalised.

Then there's also the porting aspect. At this stage, only j2nsm has the source code to jj2 but they will never give it to us so that is not an option. Since there is no source code of any sort, the only way to port or remake will be to default to experimentation or reverse engineering of some sort. In which case, it is definitely a better idea to create a "port" or remake on the PC itself; think OpenJazz2. Only when you get something that works closely enough to the original will it be worthy of porting.

Last but not least, programming on the DS. This is what I am most skeptical about. Like ahat has already shown a few times, it is almost certain that you have never programmed any DS homebrew. Perhaps you considered learning the ropes while working on the project. This might work for windows programming, which requires lots of trial and error. But DS programming is very specific, and if it is indeed your first project, you may want to consider my first option. Which is to remake on the PC first then port it to the DS.

Conclusion
Consider whether this project will be realistic. Probably start with something slightly lighter and proceed from there.

Epilogue
OK, I tried to keep this criticism as free of attacks as possible. My apologies if it sounds rather offensive, but I believe it best to correct the entire thing before it starts to go wrong. Also, NDS > PSP!!!