Human genes may often be inserted bacteria but, why does it NOT work to insert bacterial genes into humans?

Bacteria (usually E. coli in labs) can have one plasmid (or more, of course, plus other types of DNA). It's basically circular DNA. We can cut a specific region on a plasmid, because we have its genetic sequence, with a specific enzyme. Then we can attach a gene into the cut area.

Let's say your enzyme targets the sequence GATTACA - when it sees this sequence, it cuts right in the middle, so you have GAT and TACA (and the other strand). Then we attach the corresponding "letters" to the gene of interest (CTA and ATGT) and the ends come together.

Any human gene is huge. Many, many, many letters huge. So your chances of having the sequence such as "GATTACA" are fairly high. The enzyme can therefore randomly cut the middle of a human gene and make it useless. So it's not that we can't, it's just technically challenging.

That said, scientists CAN insert viral DNA into human DNA. Actually, that's how some viruses work (called retroviruses, for instance HIV). Scientists can take those viruses, insert a gene of interest (in the same way as described above), and use the virus' own "technology" to insert it into human DNA (obviously, first they remove the 'bad' virus genes and only leave the 'good' ones).