diff --git a/brozzler/worker.py b/brozzler/worker.py
index faa7081..8416ec0 100644
--- a/brozzler/worker.py
+++ b/brozzler/worker.py
@@ -249,15 +249,15 @@ class BrozzlerWorker:
 
         if not self._needs_browsing(page_headers):
             self.logger.info("needs fetch: %s", page)
-            if site.video_capture in [
+            if site.pdfs_only and not self._is_pdf(page_headers):
+                self.logger.info("skipping non-PDF content: PDFs only option enabled")
+            elif site.video_capture in [
                 VideoCaptureOptions.DISABLE_VIDEO_CAPTURE.value,
                 VideoCaptureOptions.BLOCK_VIDEO_MIME_TYPES.value,
             ] and self._is_video_type(page_headers):
                 self.logger.info(
                     "skipping video content: video MIME type capture disabled for site"
                 )
-            elif site.pdfs_only and not self._is_pdf(page_headers):
-                self.logger.info("skipping non-PDF content: PDFs only option enabled")
             else:
                 self._fetch_url(site, page=page)
         else:
diff --git a/job-conf.rst b/job-conf.rst
index d0428ca..5378bac 100644
--- a/job-conf.rst
+++ b/job-conf.rst
@@ -114,10 +114,12 @@ seeds starving out other jobs.
 +=========+==========+===========+
 | boolean | no       | ``false`` |
 +---------+----------+-----------+
-Limits capture to PDFs based on MIME type. This value will only impact
-processing of outlinks within Brozzler. Fully limiting a crawl to only PDFs
-requires an additional entry in the Warcprox-Meta header ``mime-type-filters``
-key.
+Limits capture to PDFs based on the MIME type set in the HTTP response's
+Content-Type header. This value only impacts processing of outlinks within
+Brozzler.
+
+*Note: Ensuring comprehensive limiting to only PDFs requires an additional
+entry in the Warcprox-Meta header `mime-type-filters` key.*
 
 ``seeds``
 ~~~~~~~~~