MATT_KB,
The technique really relies on UV mapping that is ‘human readable’, so that you can make sense of which bit goes where. The texture maps that you produced using the technique will have the ‘front’, ‘side’ and ‘three-quarters’ in the right positions but with a lot of extra white space; you need to merge them into a single texture map that when displayed on the model will look good.
The simplest way to work is to load the texture maps into Photoshop as different layers of a single image. Adjust layer transparency so that you can see through, then blend the layers together by carefully erasing the extra bits for each layer. You may need to use the clone tool to tidy some areas.
You might find it useful to work with the UVs as background. You can create this in ZBrush: with your model in Edit mode, create a new texture of the same size as your maps then press Tool: Texture: UV>Txr. This will create a texture with your UVs that you can export.
Finally, delete the UV background and merge the texture Layers together to form a single map.
HTH,