Single crystal diffraction pattern data coupled with a rough idea of both the size and expected crystal structure from diffractopas can all come together to help you determine the crystal structure to an extent. However, Rietveld refinement is only used when there is a clear picture of the crystal structure and you want to obtain a more precise lattice parameters, angles, atomic positions  and so on  although it is not as easy as it sounds especially when there is an overlap of diffraction maximums, but been able to obtain the lattice constant, space group and performing some indexing using the Miller Index formula is a sure way to get an accurate crystal structure with accurate lattice parameters can be placed in the database of the International union of crystallography. Remember, the size of your crystal is a major determinant of obtaining good results because you would need to orient your crystals to obtain the right patterns and also check for higher order rotations.

Hope this helps.

Yes, I agree with Mr. Bakare that XRD diffractogram gives a rough idea about the particle size and the structure of the material. Therefore, you need Rietveld refinement, to get exact crystal structure and lattice parameters of the material.To perform Rietveld refinement, you can use the full_proof software.I hope this will help you.

I guess you simply did not correctly formulate your question. Using two reflections you get access on two numbers, the position of the reflection (assuming the peak only describes a single reflection and no overlay). The peak width you can use as well as the intensity, however there are several factors which affect these properties. Concluding: we have two single numbers! This in maximum enables a determination of two lattice parameters, if the peaks are not formed by the same lattice plane since then you can only derive the lattice parameter, if the phase is cubic. For the crystal structure you at least need so many reflections as the structure is described by independent atomic ccordinates, i.e. it is mathematically impossible like the equation: x+3=y. It has infinite solutions. Even for the determination of the lattice parameter it is not recommended to use a single reflex because of experimental errors. The equation changes to (2+delta)+3=y. Delta might be small but it directly influences the resulting value of y. in order to reduce delta one uses as many as possible peaks and the upper mentioned refinement which is a "simple" least-square algorithm with the intention to vary the lattice parameter until the final deviation between experimental and theoretical curve is minimal. This has certain limitations and requires a very careful tuning, e.g. the small deviation can be caused by the background but also by an entire peak. This means, you have to prove carefully whether the refinement makes sense. A small R value is a nice indication but need some check.

Coming back to your question: I assume that you are interested whether two peaks are sufficienct to discriminate between two or more modifications of ZnO. Well, as long as you are sure that there is no other phase, and you select one or two peaks which are very phase-specific then you can do this. It is not the most convincing procedure but acceptable if you can exclude any other interpretation of these reflections. For phase identification (!), i.e. one even did not say it is chemically ZnO, in former days even only three reflections have been used. If you look in older XRD labs you possibly still find the Hanawalt (three strongest peaks) or Fink index (eight strongest peaks) which have been used to identify phases. It is clear that if you reduce the candidates to only a very few one this identification works much better. However, it assumes that nothing else exists and that you know all phases, i.e. practically the crystal structure (although in case of powder diffraction data you do not need this really since for many phases the powder diffraction data are known but no crystal structure data are available). For ZnO practically both exist. 

But I want come back to the beginning: Your question is practically incorrectly formulated. You are not looking for the crystal structure data since then I have to say, the assumption to do this with two peaks is redicules! But you do not need the crystal structure. You can do it with the lattice parameters and the resulting reflection positions. The difference between crystal structure and crystal lattice is essential and you should try to understand this reading a book about fundamentals in crystallography. In other words, you nether need single crystal techniques nor Rietveld refinement for this procedure. Both methods deliver something you were actually not asking for. It is comparable to the question: How I can come from point A to point B? One recommends you to build a subway, the other one tells you you need to build a highway. I simply recommend you to walk. It is easier since you know how to do this, and it is even faster since you do not need to build anything.

http://www.icdd.com/profile/history/Vol.1.1.1986ManualSearch.pdf

DId you get these two peaks by using copper K-alpha radiation for the X-ray powder diffraction experiment. This information is also needed when you want to identify if the ZnO belongs to the hexagonal modification or to the cubic (zinc blende) form.

Of course one has to be careful in the phase identification because a large number of papers are being written on zinc oxide materials where the characterization is less than satisfactory.  A recent comment in the March 2017 issue of Indian j Chem (section A) for one such ZnO material can be useful.