I don't think this would be practical. Quite apart from the amount of effort involved to deal with the many previous products and their various amp models, the biggest problem would be dealiong with the fact that the HD series is advertised as being a 10-fold improvement in terms of sampling detail. In other words, a conversion would have to map each bit of information into 10 bits - with absolutely no information about how to do the mapping. Imagine handling two consecutive bits of X3 information with values 1 and 2, and having to assign 10 interim values within the range. Would you assign them linearly (1.0, 1.1, 1.2, 1.3,...1.9, 2.0)? Or would you use some other algorithm (e.g. 1.0, 1.0, ...1.5, 1.5, 2.0)? No matter what you do, the result in HD terms would probably sound very different from the actual HD modeling sounds. I think it's an admirable idea, but an impractical one.
Or, to use perhaps a more familiar analogy, imagine taking a photo with your digital camera at a low resolution intended for sharing via email. That's the X3. Now imagine you want to print that photo in a 10x14 portrait. The resolution is going to be blurry, and there's nothing you can do about it because the detailed clarity simply does not exist in the data. On the other hand, if you had originally taken the photo in a hi-res format (the Pod HD) a 10x14 printout would look great, with much more visual detail and clarity.
The reason that conversion works among the Pod HD products is that they all use the same modeling technology, and the underlying data resolution is the same. It's really just a difference in mapping inputs and outputs as well as filenames, with a few manageable variances in FX options.