diacritics-neutralise has trouble when characters with modifiers are used

In Unicode you can make a character like "ý" as either a single standalone character U+00FD or a 'y' followed by an accent, U+0079U+0301.

diacritics-neutralise doesn't really even try to mediate between these.

A relatively simple possible solution is to use the normalize() method on javascript strings.

function removeDiacritics(s) {
  s = s.normalize();
  for (var i = 0, l = diacriticsMap.length; i < l; i++) {
    s = s.replace(diacriticsMap[i].letters, diacriticsMap[i].base);
  }
  return s;
}

I've done this to my local copy of the code. I don't know yet if there will be any ill effects, but I suspect not, since both the search and target strings will be normalized.

Locate the diacritics-neutralise implementation and its removeDiacritics function. Check how JavaScript string normalization interacts with the existing diacritics map, then verify behavior for both composed and decomposed forms of the accented character. Done means equivalent Unicode forms are handled consistently without breaking existing replacements.