Back in 1649, when Descartes formulated the concept of a reflex, it was used to describe lower animals, to support his notion that they were automata without a mind on their own. The word reflex originated from the "reflection" of the sensory input into a response. The physiological backgrounds were not yet known.

In the early 19th century, Hall narrowed the definition of a reflex to be a "involuntary action of a muscle or gland in response to the stimulation of a receptor neurone which does not depend on the existence of consciousness".

Sherrington, in 1904, narrowed the definition further, introducing the concept of the "reflex arc", a hardwired pathway between receptors and effector muscles.

Pavlov later did a strange thing and widened the definition, which is the source of some confusion. Nowadays, Pavlov's Conditioned Reflexes are mostly called Conditioned Responses, to avoid this exact confusion. NB: there are people arguing that conditioned is a mistranslation, and that it should mean conditional, which makes more sense in the context.

Pretty much every current publication uses the term "reflex" for involuntary, inborn responses, detected by receptors, transported to the spinal cord by afferent nerve fibers, processed there, and the motor signal sent through the efferent nerve fibers to the effector muscles. I only found the term "conditioned reflex" in papers discussing Pavlov's experiment.