Since Word 2013, Word has had the ability to directly open pdf document. However, some pdf documents (from scanners mostly) are really pictures of text rather than the text itself. Just as you can't look under the hood of a picture of a car, you can't edit a picture of a document.

In that case, you need to convert the picture to text. This is a process known as optical character recognition. This is built into Adobe Acrobat and is also in Office OneNote. Most scanner software comes with an OCR component as well. Once you have text, you can edit it with Word.

If you simply want to write on the document (but not in it) you can add a Text Box floating on top of the document layer, whether or not it has been put through the OCR process.

Word documents that have been saved as PDF will not need the OCR process, they retain their text, although not all of their Word structure and formatting.

Finally, documents converted from pdf (or really any other format) to Word can be tough to edit because the conversion process never has a one-to-one matching of how formatting is done. This means that a converted document will seldom be formatted in Word in a way that makes sense. An example is multiple section breaks to change margins, where in Word you would simply change the paragraph indent. Margins and Indents in Word