Generally speaking, the idea of "progress" in art is an untenable one. Styles change, new ideas come along, different art is created, but the old art retains its value. Anyone interested in early music, or classical music in general, already believes this to some degree, and may even believe that later art is decadent. Here we take the view that the music of each era has its own strengths, and that each era is fundamentally as good as any other. That is the polite thing to do, even if individuals will always have their own stylistic preferences.
Similar comments apply to interpretations of music. Different decades have different fashions for musical interpretation, and for the most part, preferences are a matter of taste. The EM FAQ must prioritize HIP interpretations of e.g. 18th century music, but the truth is that such preferences are largely matters of taste, and other interpretations can have arguably equal merit. Leading interpretations in whatever style are all based on a deep appreciation for the music. When it comes to earlier music, Medieval & Renaissance music, this sort of "objective equality" between interpretations of similar stature is not entirely true.
Interpretive choices made by performers from decade to decade retain a large component of fashion, but there are also a couple of objective reasons that interpretations of Medieval & Renaissance music have improved, at least in some sense: 1) Our understanding of what the old musical notation means has improved. Real misunderstandings have been rectified regarding the marks on the paper; ideas on choosing accidentals in polyphony and rhythms in unmeasured music have developed; the basic theory, tunings, and ensemble constitutions have been uncovered. In some cases it has been a matter of taking early theorists seriously (e.g. Pythagorean tuning), instead of believing the music must have sounded more like later music. 2) Performers have had more opportunity to hear the music as performed by others, to internalize it and gain a personal feeling for it. No matter what one's interpretive stance on Beethoven might be, any performer has had ample opportunity to hear Beethoven's music performed, for it to seem real, for an interpretation to concentrate fully on expression rather than on feeling one's way in the dark.
Having never been able to hear the music before, working with nothing but strange marks on a page, the work of the pioneering performers of early music is that much more impressive. The value of their work is not really diminished by the fact that more discoveries have been made since they embarked on their landmark performances. One thing these performances will always have is the excitement of bringing something back to life, and so they have a uniqueness that the performances of today will cease having as soon as there are performances of tomorrow. The pioneering performers did things in their interpretations that are now considered incorrect, that they themselves would not or do not do now, and so performances of early music have improved in that sense. It is also true that simply knowing does not guarantee a good interpretation, as there remain issues of communication and passion and talent. One can even say that the latter issues override correctness, and they can, but it is also difficult to embrace a passionate interpretation in which e.g. every A should be Ab.
How long can interpretations continue to improve in this sense? In much of this repertory, we now (apparently) have a full understanding of basic technical facts of note & rhythm. In some cases, this understanding has been secured only in the past few years (e.g. accidentals & relative tempi in Franco-Flemish polyphony) and not yet fully transmitted. In a diminishing number of cases (e.g. Ars Subtilior notation), that understanding is not yet secure. Once we understand the music on this level, as we do Beethoven's, the possibilities for any "objective" improvement diminish. There will remain other issues of sonority and ensemble, but consensus on some of these aspects is also being reached. At some point, discoveries and debates will become increasingly trivial, and largely indistinguishable from fashion. At some point, performers of this music will have heard it as children as much as they might have heard Beethoven's, and will be able to feel it just as deeply. At that point, changes in early music performance will essentially be dictated by fashion alone, or rather the personal visions of individuals equipped with a full technical understanding of the music.
The above should not be taken to suggest that we will ever know everything, but rather only that we will exhaust our available information on the subject. In some ways, this is when the true fun begins, because some of the unanswerable questions surrounding early music allow for a great deal of performer creativity. Some of the earlier performances of Medieval & Renaissance music, however, whatever their merits, adopt choices which are simply false, even on something as basic as the relative notes themselves. One can consider this a matter of taste anyway, but it is a treacherous matter best undertaken with a full knowledge of the subject. The departures in earlier performance are generally not deliberate, even if deliberate departures from HIP can be compelling in their own way. In many repertories, performances remain a major part of an ongoing process of discovery.
Finally, it should be noted that recommending a recording of a particular piece of music which is fully up to date on scholarly issues, is performed with feeling & talent, and is available for purchase, is frequently impossible. What the novice listener receives may well involve a compromise on one or more fronts.