Yes, it has a different image processing algorithm that will find boundaries at a sub-pixel level. This can really help dense 1D barcodes.
False positives do happen, and in particular it is easy to misread UPC-E codes in a barcode. Am I right that this is what is happening?
The Code 128 check digit doesn't help in the sense that it's not mis-reading a Code 128 code, but a phantom barcode of another type.