Where the text says:
If a universal-character-name does not correspond to any character in ISO/IEC 10646 [ ... ].

The note indicates:
ISO/IEC 10646 code points are within the range 0x0-0x10FFFF, inclusive.

I am not certain what the text is meant to say.
Is the intention merely that the numeric value shall be within the UCS codespace (0x0-0x10FFFF, inclusive)?
Is the intention that the code point shall be one of those that are assigned to characters by ISO/IEC 10646 (in which case, the additional limitation on surrogate code points is redundant)?
Is the intention that the code point shall not be one whose basic type is the noncharacter type?

Also, the paper should probably including something like the following:
Modify in [lex.string],
<del>Within char32_t and char16_t string literals, any universal-character-names shall be within the range 0x0 to 0x10FFFF.</del>