WD04 lines 259-261 describe the way that 16-bit data values are presented. The text is correct, however it appears in section 2.1 which is about the Fixed Header, and that header does not contain any 16-bit words. Also I think I am right in saying that each place in the spec where a 16-bit value does occur, there are words that state the byte order. So these words aren't needed in 2.1 and should be removed. You could consider putting them in Section 1, and saying that they apply to all 16-bit integer quantities anywhere in the spec (in which case you could argue that there's no need to repeat them each time the spec mentions a 16-bit quantity)