[lex.pptoken] p2 says:

Preprocessing tokens can be separated by whitespace; this consists of comments ([lex.comment]), or whitespace characters (U+0020 space, U+0009 character tabulation, new-line, U+000b line tabulation, and U+000c form feed), or both.
As described in [cpp], in certain circumstances during translation phase 4, whitespace (or the absence thereof) serves as more than preprocessing token separation. Whitespace can appear within a preprocessing token only as part of a header name or between the quotation characters in a character literal or string literal.

"can" in this sentence serves as a weak tone. The intent is, whitespace definitely separates preprocessing tokens except for some special cases. The improvement is:

Except that the whitespace that appears within a preprocessing token only as part of a header name or between the quotation characters in a character literal or string literal, a whitespace separates preprocessing tokens.