When you register a narrow-field image against a wide-field image, the resulting aligned image is zero everywhere except for the intersection of both input images. This is just the expected result. If you see the registered image "white", this typically happens because you apply an incorrect STF, or STF AutoStretch computed for a different image. It's difficult to say anything useful without taking a look at a screenshot, an example image, or something concrete.
If you want to integrate these images, just feed the ImageIntegration tool with them, as usual. Leave the clip low range and range low parameters enabled and zero (default values), respectively, and it will work without problems. Or you can integrate both sets independently (narrow and wide field apart) and combine the two resulting images with PixelMath using something like max(A,B) or a similar function.
