This monograph by Ronald O. Richards is based on his PhD thesis at UCLA from 2001. By using the earliest Slavic loanwords in Hungarian, Richards seeks to reconstruct the Slavic dialect that was spoken in Pannonia when the Magyars conquered the area in AD 896.
Of course, first one has to determine what the "earliest" loanwords were. Richards describes his methodology in some depth, mainly that the presence of vowel+nasal sequences in Hungarian for Common Slavonic nasal vowels indicates an early date, because the nasal vowels disappeared in the Slavic languages of the area by the beginning of the second millennium. There are some other criteria based on Hungarian-internal developments. Once Richards has assembled a set of Slavic loans that look to be early, he compares them to early Slavic dialects of the area and finds that Pannonian Slavic was a South Slavic language, closest to Proto-Serbo-Croatian and not closely related to Proto-Slovene.
The biggest downside of this book is the way it is made: a cheaply typeset (early word processor?), printed (computer printer?) and bound (unbending glue binding) paperback. It won't lie anything close to flat, and you might just find that it disintegrates after a single reading.
Besides that, I have a few minor quibbles with the content: Richards doesn't gloss a lot of the words, either Hungarian or Slavic, and no reader is likely to know all of the words here (especially when some of the Hungarian words are archaic and not found in modern dictionaries). I think he also brushes off the prospect that Old Hungarian could have drawn Slavic loanwords from East Slavic when the Magyars were still in what is now Ukraine.