In python-docx there's no direct access to images in the document structure. To extract text, tables, or images in the original order, you have to manually parse the XML tree (doc.element.body.iter()), checking for tags like w:p for paragraphs, w:tbl for tables, and w:drawing or w:pict for images.