What's the difference? I would all encourage you to read this book:
Richard Muller is fantastic scholar. He was also the doktor-vater of Rev. Dr. Ben Mayes of Johann Gerhard translation fame.
In this book, Muller describes the genesis of the loci method. The idea originally came from Rudolf Agricola in the 15th century. Agricola made the somewhat novel argument that rhetoric should proceed logic. In other words, to know what's logical, you have to first hear a good argument and then abstract from it that what's rational.
In order to work out this method, he used as his model "the good arguments" of classical rhetoricians. In that everyone respected, let's say Cicero as a rhetorician, clearly his arguments must be good. He thought that were specific spots in their books were you could find the clear and good arguments could be the sources or good rhetoric and therefore good logic.
In dealing with the interpretation of texts, Erasmus took over Agricola's method as a form of rhetorical analysis. Melanchthon, being the Humanist that he was, took over this method of rhetorical analysis and applied it to the Bible and then parlayed it into the method of writing dogmatic theology. From there is spread to the Lutheran and Reformed scholastics.
How does this work? Melanchthon used first the book of Romans (in the original Loci Communes of 1521) and then later the creed to form a kind of skeleton for his dogmatic work. Then he would discuss each article of the faith in its own separate loci.
Each article of the faith had it's own "sedes doctrina" that is "seat" in Scripture where a doctrine could be found. These were the clear passages in Scripture. For example for the Trinity you could look at "In the beginning was the Word and the Word was with God and the Word was God." Or with creation you could read "In the beginning God made the heavens and the earth." Both passages (if you know Biblical languages) are very grammatically simple and therefore very clear. The theologian then looks at those passages and engages in exegesis to fill out the exposition of the doctrine in the loci within the structure of the dogmatic work. There may also be unclear passages where the doctrine is discussed, but those are illuminated by the clear ones. All are coordinated with the analogy of the faith, which centers on the doctrine of justification by faith. Nevertheless, each doctrine has its own sedes and is not dependent on other articles for its articulation.
Now, here's the advantage of this from the perspective of Preus. Since each loci is investigated independently of the others and the order of articles doesn't matter either (according to Muller interpretations of Protestant scholasticism have wrongly thought that the order mattered, it didn't. It usually based on the creed or in the case of the first Loci Communes, the book of Romans) then there is a greater allowance for paradox. In a word, because each article can be expounded separately from the others, they can simply stand as they are. So, if justification and universal grace don't logically cohere with what the Bible says about predestination, so be it! You just let the doctrine stand and don't try to resolve the paradox. For this reason, Preus thinks that the loci method allows for greater appreciation paradox, opposed to a method where all articles are deduced from central articles (like justification or election). In that case, one will start ironing out the paradoxical unity of different articles of the faith through human reason and thus distort the teaching of Scripture. Barth, though he says he prefers the loci method, certainly tends to do the later through his Christomonism.
Now is this exactly true? Well, theoretically it should be thus. The problem is that it doesn't appear to do so in practice. For one thing, the teaching of double predestination found in the Reformed tradition was developed by theologians using this method and as we are well aware, this teaching is pure rationalism. The same could go for Lutheran theologians who developed inuitu fidei (Leonard Hutter probably being the first). So, despite the imperative to let paradox stand, it doesn't always work out.
Of course, if this were consistently followed as a dogmatic method (as I think Gerhard does) then the result should be as Preus claims. For that reason, I think that the loci method has much to offer contemporary theology.